-
An Experimental Evaluation of Fixed and Fluidized
Beds of Zeolite 13X for the Application of Compact
Thermal Energy Storage
by
Dylan A. Bardy, B.Eng. (Carleton University, 2016)
A thesis submitted tothe Faculty of Graduate and Postdoctoral
Affairs
in partial fulfillment of the requirements for the degree of
Master of Applied Science in Sustainable Energy
Department of Mechanical and Aerospace EngineeringCarleton
University
Ottawa, OntarioAugust, 2018
©CopyrightDylan A. Bardy, 2018
-
Abstract
For thermal energy storage technologies based on physical
adsorption to become
a commercially viable option in the future, particular
advancements in the research
and development of the system’s components are required to
complement existing re-
search in advanced materials. To investigate the application of
fluidization as a solid-gas
contacting method for low-temperature thermochemical energy
storage, a bench-scale
adsorption-based TES system was designed, constructed,
instrumented, and commis-
sioned. In demonstrating this technology, the objective of this
research was to obtain
thermodynamic data for the adsorption of water vapour onto
zeolite 13X under flu-
idization to evaluate fluidized beds as potential reactor or
adsorber designs. Multiple
adsorption experiments were performed on samples of an 8x12 and
60x65 mesh zeolite
13X molecular sieve, comparing the effects of air flow rate and
concentration of water
vapour on the breakthrough and temperature lift on the energy
density of fixed and flu-
idized adsorbent beds. Variation of the air flow rate from 10 to
30 L/min had little effect
on the amount of thermal energy released by the adsorption of
water onto each sample
due to equilibrium; however, the added volume of the bubbles in
the fluid bed reduced
its energy density significantly at higher flow rates. The
concentraion of water vapour
at the inlet or relative humidity of the air was shown to be one
of the most significant
parameters for controlling the delivery temperature of each bed.
It was found that, the
discharge rate of the fluid bed increased by 4 W per 20%
increase in RH, compared to a
6 W per 20% RH increase for fixed beds. Tests were performed to
determine the effects
of fluidization and the use of a vacuum pump on the regeneration
of zeolite 13X. The
results of the experimental data are were then considered in
scaling the adsorbent beds
to meet a range of heating loads.
i
-
Acknowledgements
I would like to thank my supervisor, Dr. Cynthia Cruickshank,
for her mentorship and
support over the last three years; this experience has been a
highlight of my career and
I hope that our paths cross again. I would also like to thank
Dr. Handan Tezel for her
co-supervision and expertise in guiding my research.
To Bill Wong for his interest in my project and for facilitating
our industrial partnership
on behalf of Leidos Canada. I must also thank Leidos Canada and
the Natural Sci-
ences and Engineering Research Council of Canada (NSERC) for
their financial support
through the Engage, and Collaborative Research and Development
Grants (No. 492298
and 500831).
Big thank you to the guys in the machine shop, Alex Proctor, Ian
Lloy and Kevin Sang-
ster for the help and training they provided me in the machining
of the parts for the
adsorption column.
To my colleagues in the lab, Chris, Brock, Nina, Jordan and
Tyler, it was a pleasure to
work alongside you all. I would like to specifically thank Alex
Hayes for his friendship,
contributions to my experimental setup, and all of the
belays.
To my family, Mom and Dad; Grandma and Grandpa; and Kelsey and
Sarah for their
love and unwavering support.
Finally, to Cam for everything over the last year. I look
forward to what’s next.
ii
-
“If the fool would persist in his folly he would become
wise.”
– William Blake (Proverbs of Hell, 1790)
In loving memory of Zachary Burgoyne (1992 - 2017).
iii
-
Table of Contents
Abstract i
Acknowledgements ii
Table of Contents iv
List of Tables vi
List of Figures vii
Nomenclature x
1 Introduction 1
1.1 Motivation . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . 1
1.2 Background . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . 2
1.3 Research Objectives . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . 9
1.4 Contribution of Research . . . . . . . . . . . . . . . . . .
. . . . . . . . . 9
1.5 Organization of Research . . . . . . . . . . . . . . . . . .
. . . . . . . . . 10
2 Literature Review 11
2.1 Global Status of the Technology . . . . . . . . . . . . . .
. . . . . . . . . 11
2.2 Systems for Thermochemical Storage . . . . . . . . . . . . .
. . . . . . . 17
2.3 Fundamentals of Fluidization . . . . . . . . . . . . . . . .
. . . . . . . . 19
2.4 Performance Metrics . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . 23
2.5 Comparative Studies . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . 25
2.6 Research Direction . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . 36
iv
-
3 Experimental Design 37
3.1 The Experimental Setup . . . . . . . . . . . . . . . . . . .
. . . . . . . . 37
3.2 Material Preparation . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . 46
3.3 Procedure . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . 47
3.4 Calculation of Various Adsorption Parameters . . . . . . . .
. . . . . . . 50
4 Results & Discussion 52
4.1 Variation of Air Flow Rate . . . . . . . . . . . . . . . . .
. . . . . . . . . 52
4.2 Variation of Relative Humidity . . . . . . . . . . . . . . .
. . . . . . . . 59
4.3 Adsorption-Regeneration Cycling . . . . . . . . . . . . . .
. . . . . . . . 64
4.4 In-Situ Regeneration Under Partial Vacuum . . . . . . . . .
. . . . . . . 66
4.5 Scaling Considerations . . . . . . . . . . . . . . . . . . .
. . . . . . . . . 69
5 Conclusions and Future Work 73
5.1 Conclusions . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . 73
5.2 Future Work . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . 74
References Cited 76
Appendices 82
A Adsorption Column Drawings 83
B Manufacturer Specifications & MSDS for Zeolite 13X 96
C Fluidization Velocity Calculations 102
D Calibration &Uncertainty Analysis 105
D.1 Component Calibration and Uncertainty . . . . . . . . . . .
. . . . . . . 105
D.2 Overall System Uncertainty Analysis . . . . . . . . . . . .
. . . . . . . . 107
v
-
List of Tables
1.1 Reversible solid-gas reactions considered for thermal energy
storage [6]. . 5
2.1 Key achievements by Task 42/Annex 29 Working Group A1. . . .
. . . . 14
2.2 Key achievements by Task 42/Annex 29 Working Group A2. . . .
. . . . 15
2.3 Key achievements by Task 42/Annex 29 Working Group A3. . . .
. . . . 15
2.4 Key achievements by Task 42/Annex 29 Working Group B. . . .
. . . . . 16
2.5 Key achievements by Task 42/Annex 29 Working Group C. . . .
. . . . . 17
2.6 Characterization of adsorbents for thermochemical storage. .
. . . . . . . 27
2.7 Lab-scale testing of open-systems. . . . . . . . . . . . . .
. . . . . . . . . 29
2.8 Lab-scale closed-system thermochemical projects. . . . . . .
. . . . . . . 31
2.9 Open pilot-scale thermochemical projects. . . . . . . . . .
. . . . . . . . 33
2.10 Closed pilot-scale thermochemical projects. . . . . . . . .
. . . . . . . . . 35
3.1 Volume of fluid bed in column at each flow rate. . . . . . .
. . . . . . . 43
3.2 Configuration of I/O modules for the NI cRIO 9204. . . . . .
. . . . . . . 45
3.3 Labels of samples for tests varying flow rate. . . . . . . .
. . . . . . . . . 48
3.4 Labels of samples for tests varying inlet relative humidity.
. . . . . . . . 48
C.1 Estimated terminal velocities and flow rates for a 210 µm
particle. . . . 104
vi
-
List of Figures
1.1 Classification of modes for thermal energy storage . . . . .
. . . . . . . . 3
1.2 The desiccant cycle . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . 7
1.3 Thermal energy density of systems compared to mode, adapted
by [18]. 8
2.1 Organization of IEA-SHC Task 42/ECES Annex 29 . . . . . . .
. . . . . 13
2.2 Open-system concept for thermochemical energy storage. . . .
. . . . . 17
2.3 Integrated vs. external adsorption systems . . . . . . . . .
. . . . . . . . 18
2.4 Liquid-like behaviour of gas fluidized beds . . . . . . . .
. . . . . . . . . 19
2.5 Balance of forces acting on fluidized solid particle. . . .
. . . . . . . . . 20
2.6 Fluidization regimes of gas fluidized beds . . . . . . . . .
. . . . . . . . . 21
3.1 The experimental setup . . . . . . . . . . . . . . . . . . .
. . . . . . . . . 37
3.2 The setup configured for adsorption experiments. . . . . . .
. . . . . . . 38
3.3 The setup configured for in-situ regeneration. . . . . . . .
. . . . . . . . 39
3.4 The setup configured for in-situ vacuum regeneration. . . .
. . . . . . . . 40
3.5 Cross section of the adsorption column. . . . . . . . . . .
. . . . . . . . 41
3.6 Dimensions of fixed bed for calculating volume (units in
mm). . . . . . . 42
3.7 Dimensions of fluid bed for calculating volume (units in
mm). . . . . . . 43
3.8 A 25 g sample of 8x12 zeolite in a fixed bed and 60x65
zeolite fluidized
at 25 L/min . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . 44
3.9 Response of MFCs to change in inlet relative humidity
signal. . . . . . . 46
3.10 Subtraction of areas for calculating mass of water
adsorbed. . . . . . . . 50
4.1 Concentration breakthrough of the outlet for the fixed and
fluidized bed
for a flow rate of 10 - 30 L/min. . . . . . . . . . . . . . . .
. . . . . . . . 53
vii
-
4.2 Illustration of MTZ Concept in a fixed adsorbent bed. . . .
. . . . . . . 54
4.3 Mass transfer zone of fixed and fluidized bed with flow
rate. . . . . . . . 55
4.4 Mass of water adsorbed by the fixed and fluidized beds . . .
. . . . . . . 56
4.5 Heat released by 25 g of zeolite 13X in the fixed and
fluidized beds. . . . 57
4.6 The storage capacity and energy density of the fixed and
fluidized beds. . 58
4.7 Concentration breakthrough of the outlet for the fixed and
fluidized bed
for 30 - 70% RH at inlet. . . . . . . . . . . . . . . . . . . .
. . . . . . . 59
4.8 Outlet temperature lift of the fixed and fluidized bed. . .
. . . . . . . . . 60
4.9 Rate of heat and specific power discharged by the fixed and
fluidized beds. 62
4.10 Cumulative energy released by the fixed and fluidized beds.
. . . . . . . . 63
4.11 The system temperatures and outlet relative humidity for a
typical re-
generation run. . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . 64
4.12 The average temperature lift and discharge rate of cycled
fixed and flu-
idized beds during adsorption. . . . . . . . . . . . . . . . . .
. . . . . . . 65
4.13 The average energy released per cycle and cycle energy
densities. . . . . . 66
4.14 Average absolute pressure of the column for regeneration
under partial
vacuum and atmospheric conditions. . . . . . . . . . . . . . . .
. . . . . 67
4.15 Outlet temperatures and relative humidity of adsorption
column for re-
generation under partial vacuum and atmospheric conditions. . .
. . . . . 68
4.16 Temperature lift and energy released for regeneration under
partial vac-
uum and atmospheric conditions. . . . . . . . . . . . . . . . .
. . . . . . 69
4.17 The total storage mass and volume of zeolite required based
on annual
heating load and a floor area of 100 m2. . . . . . . . . . . . .
. . . . . . . 70
4.18 Adsorbent bed volume of the fixed and fluidized bed, based
on a floor
area of 100 m2. . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . 71
A1 Adsorption Column Assembly . . . . . . . . . . . . . . . . .
. . . . . . . 84
A2 Inlet Sub-Assembly . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . 85
A2.1 Collar for Acrylic Tube . . . . . . . . . . . . . . . . . .
. . . . . . 86
A2.2 Wing Nut Clamp . . . . . . . . . . . . . . . . . . . . . .
. . . . . 87
A2.3 1-1/8” Straight Quick-Clamp Adapter . . . . . . . . . . . .
. . . 88
A2.4 1-3/4” Straight Quick-Clamp Adapter . . . . . . . . . . . .
. . . 89
viii
-
A2.5 Quick-Clamp Gasket . . . . . . . . . . . . . . . . . . . .
. . . . . 90
A2.6 200x200 Mesh Disc . . . . . . . . . . . . . . . . . . . . .
. . . . . 91
A2.7 Inlet & Outlet Quick-Clamp Press Fit . . . . . . . . .
. . . . . . 92
A2.8 Sloped Insert . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . 93
A3 Acrylic Column . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . 94
A3 Outlet Sub-Assembly . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . 95
ix
-
Nomenclature
Abbreviations
ACH Air Changes per Hour
BTES Borehole Thermal Energy Storage
CJC Cold Junction Compensation
DHW Domestic Hot Water
DLSC Drake Landing Solar Community
FSO Full Scale Output
IEA International Energy Agency
MFC Mass Flow Controller
MSDS Material Safety Data Sheets
P&ID Piping and Instrumentation Diagram
PID Proportional - Integral - Derivative
RH Relative Humidity
RTD Resistance Temperature Detector
SLM Standard Liters per Minute
SOC State of Charge
TCM Thermochemical Material
x
-
TES Thermal Energy Storage
VI Virtual Instrument
Greek
∆ total change in variable
δ change in variable per unit of time
ω absolute humidity (gair/kgH2O)
φ relative humidity (%)
ρ density (kg ·m−3)
d diameter
General
Q̇ rate of heat discharge (kW)
V̇ volumetric flow rate (m3 · s−1)
cp specific heat capacity (kJ · kg−1 ·K−1)
M molar mass (kg ·mol−1)
P pressure (kPa)
Q thermal energy (kJ)
T temperature (◦C)
t time (s)
Subscripts
a air
ads adsorption
b breakthrough
xi
-
d discharge
e exhaustion
g gas
r reaction
s solid
tot total
w water
ws saturated water vapour
xii
-
Chapter 1
Introduction
1.1 Motivation
It is well recognized within the sphere of clean-tech that
economically viable energy
storage technologies are required to realize the full potential
of renewable energy by
correcting the mismatch between variable demand and intermittent
supply. Compared
to more established methods of energy storage (namely mechanical
and electrochemical),
thermal energy storage (TES) technology has received less
attention because of the lower
financial value associated with heat. High-grade forms of energy
capable of performing
mechanical work (i.e., electricity and fossil fuels) are the
most costly to produce, whereas
low-grade thermal energy (< 250◦C) has virtually no
commercial value [1]. Economics
from the perspective of exergy (considering second law
thermodynamic analysis) should
therefore discourage the use of high-grade energy from
combustion or electricity for
low-grade end-use applications such as space heating and
domestic hot water (DHW);
however, more than 8.5 million homes that make up Canada’s
single-detached residential
building stock are heated by baseboard electric heaters and the
combustion of natural gas
[2]. While these two conventional methods of heating can boast
high energy efficiencies,
they produce greenhouse gases, are subject to increasingly
volatile energy prices, and
rely on a distribution infrastructure that is both strained and
aging. New and energy
efficient space and water heating technologies are integral to
reducing energy usage in
Canadian residential buildings, where conventional heating
technologies account for over
80 percent of energy use in this sector [2].
1
-
As a renewable source of heat (i.e., naturally replenished
within a human timescale),
solar thermal energy can be collected in the range of 300 – 900
kWh/m2/year, which
could meet the 110 GJ heating load of a typical Canadian home
with 30 – 100 m2
of solar thermal collector area [3, 4]. While most Canadian
cities experience sufficient
insolation for solar thermal technology between the spring and
fall equinoxes, sizable
seasonal storage capacity is required for the winter months.
North America’s first seasonal solar thermal storage system for
the Drake Landing
Solar Community (DLSC) of Okotoks, Alberta, stores solar thermal
energy as sensible
heat in a borehole thermal energy storage (BTES) system. In
addition to requiring
extensive drilling, this BTES system amounts to just over 34,000
m3 of underground
storage volume for 52 homes, which limits the feasibility of
implementing similar systems
exclusively to the new construction market [5]. Therefore, to
increase the penetration
of solar thermal heating in Canada and include the potential for
retrofit applications,
more compact technologies for TES with higher energy densities
need to be developed.
Thermal energy storage through solid-gas thermochemical
processes has been demon-
strated as a more compact mode of thermal energy storage than
sensible or latent TES.
International research has identified promising material
candidates for this technology
through investigations of multiple lab and pilot-scale systems
of a packed-bed type. Flu-
idization is another method of solid-gas contacting that is used
extensively in industry
for its noted advantages of higher heat and mass transfer;
however, fluidized beds for
low-temperature thermochemical storage have only been considered
conceptually in the
literature. This work was therefore motivated to evaluate the
use of fluidized beds as a
component of thermochemical TES systems through experiments at
the bench-scale.
1.2 Background
As classified in Figure 1.1, three modes are considered for the
storage of thermal
energy: sensible, latent, and thermochemical heat. An
understanding of the principles of
each mode is required to appreciate the challenges that TES
technologies must overcome
to be a viable option in the future.
2
-
Thermal Energy Storage
Sensible Latent Thermochemical
Liquid Solid
Water Rock
Solid-Liquid Liquid-Gas
Melt Vapourize
Sorption Reaction
Solid-Gas Solid-LiquidAdsorption Absorption
Figure 1.1: Classification of modes for thermal energy storage,
adapted from[6].
1.2.1 Sensible Heat
Sensible heat refers to thermal energy stored in the change in
temperature of a
material (generally in the liquid or solid phase) [7]. The
amount of energy that can be
stored in a sensible TES system is limited by the specific heat
capacity and the density
of the material:
Q = mcp∆T (1.1)
Thermal energy, Q, is therefore stored in a material of mass m
with a specific heat
capacity cp as it experiences a change in temperature, ∆T .
Liquid water for example can
absorb 58.1× 10−3 kWh of heat per liter, providing an energy
density of 58.1 kWh/m3
for a ∆T of 50◦C. Realistic energy densities must also factor in
the added volume of
insulation to minimize heat losses. The energy density of a
1,000 L (1 m3) cylindrical
buffer tank for example, is reduced to 36 kWh/m3 if its
effective volume is increased to
1.6 m3 with the addition of just 10 cm of insulation [8].
Sensible TES consisting of tanks of water are however widely
used, especially for
diurnal TES systems, where they are cycled daily. Liquid water
systems can also be
modular and extended beyond the volume of a single tank as
demonstrated by Cruick-
shank and Harrison [9] in their evaluation of a multi-tank TES.
Smaller water tanks such
as these have been used commercially as buffers for meeting peak
hot water demand from
boilers and renewable heating technologies in domestic
systems.
3
-
1.2.2 Latent Heat
Thermal energy stored in the change in phase of a material at
constant temperature
is referred to as latent heat [7]. The phase change materials
(PCM) considered are
generally those that can transition from solid to liquid, or
from liquid to vapour, and
therefore the specific heat of solidification/fusion or
vapourization and the temperature
at which the phase change occurs are of design importance. The
energy stored in a PCM
undergoing a change in phase between a liquid (cp,l) and a solid
(cp,s) is given by:
Q = m[cp,s(Tm − Ti) + ∆hm + cp,l(Tf − Tm)
](1.2)
where ∆hm is the latent heat of fusion, and Tm, Ti, and Tf are
the melting, initial and
final temperature of the PCM, respectively. PCMs can be
contained in conventional
building elements or ecapsulated as self-contained elements.
Wert et al. [10] have com-
bined the two methods by incorporating plates of an encapsulated
organic PCM with an
energy density of 50 - 55 kWh/m3 into the duct work of an
experimental setup represent-
ing a commercial HVAC system. Other demonstration projects have
been deployed and
R&D efforts are focused on improving thermal stability to
enable commercialization.
1.2.3 Thermochemical Heat
Thermochemical heat is the product of reversible chemical
reactions and physical
adsorption. Most of these processes take place between a solid
thermochemical material
(TCM) and a gas in a component that is typically referred to as
a reactor.
Reversible Chemical Reactions
Thermochemical storage gets its namesake from the reversible
chemical reactions
that make up this mode of thermal energy storage. In theory,
waste heat from an
industrial process, or solar thermal energy can be stored and
released in a loss-free
manner by means of a bimolecular reaction:
A(s) + B(g)discharge−−−−−⇀↽−−−−−storage
C(s) + ∆Hr (1.3)
4
-
The forward reaction of Equation 1.3 is exothermic, and heat is
a product of the
association of species A and B to form species C and the heat of
reaction, ∆Hr. Heat is
therefore stored in the reverse endothermic reaction of
dissociating species C into species
A and B. The reversible chemical reactions that are considered
most in the literature
for thermochemical storage have been summarized by Letcher [6]
in Table 1.1.
Table 1.1: Reversible solid-gas reactions considered for thermal
energy stor-age [6].
Type of Reaction Reaction Temperature [◦C]
Dehydration
of salt hydrates
MgSO4·7H2O −−⇀↽−− MgSO4·H2O + 6H2O
MgCl2·6H2O −−⇀↽−− MgCl2·H2O + 5H2O
CaCl2·6H2O −−⇀↽−− CaCl2·H2O + 5H2O
CuSO4·5H2O −−⇀↽−− CaCl2·H2O + 4H2O
CuSO4·H2O −−⇀↽−− CuSO4 + H2O
100 - 150
100 - 130
150 - 200
120 - 160
210 - 260
Deammoniation
of ammonium chlorides
CaCl2·8NH3 −−⇀↽−− CaCl2·4NH3 + 4NH3CaCl2·8NH3 −−⇀↽−− CaCl2·2NH3
+ 2NH3MnCl2·6NH3 −−⇀↽−− MnCl2·2NH3 + 4NH3
25 - 100
40 - 120
40 - 160
Dehydration
of metal hydrides
MgH2 −−⇀↽−− Mg + H2Mg2NiH4 −−⇀↽−− Mg2Ni + H2
200 - 400
150 - 300
Dehydration
of metal hydroxides
Mg(OH)2 −−⇀↽−− MgO + H2O
Ca(OH)2 −−⇀↽−− CaO + H2O
Ba(OH)2 −−⇀↽−− BaO + H2O
250 - 350
450 - 550
700 - 800
Decarboxylation
of metal carbonates
ZnCO3 −−⇀↽−− ZnO + CO2MgCO3 −−⇀↽−− MgO + CO2CaCO3 −−⇀↽−− CaO +
CO2
100 - 150
350 - 450
850 - 950
Of the reactions listed in Table 1.1, those that involve the
dehydration of salt hy-
drates and metal hydroxides are the most appropriate for thermal
energy storage systems
for buildings with water vapour and heat being the only
products. The deammoniation
of ammonium chlorides and dehydration of metal hydrides would
require additional com-
ponents for separating and collecting ammonia and hydrogen gas
which pose a hazard
to human health.
5
-
Physical Adsorption
Thermal energy can also be stored in the adsorption of gas
molecules onto the
surface of solid adsorbent materials, where the component that
facilitates this process is
referred to as the adsorber. Adsorption as a result of only van
der Waals interactions and
weak electrostatic forces is referred to as “physical
adsorption” or “physisorption”, and
involves the formation of multiple molecular layers of adsorbate
on the solid adsorbent.
Most the of literature on adsorption for the application of
thermal energy storage refers
to the adsorption of water vapour onto fixed beds of adsorbent
according to Equation 1.4:
A(s) + nH2O(g)discharge−−−−−⇀↽−−−−−storage
A(s)·nH2O(s) + ∆Hads (1.4)
In the forward step of Equation 1.4, multiple layers of the
adsorbate (gaseous water)
are formed on the surface of a solid adsorbent, A. This step is
exothermic because the
molecules of adsorbate are more stable on the surface of the
adsorbent than in the free
phase [11]. It is this heat of adsorption, ∆Hads, that can be
used for heating space or
DHW. During desorption, or the reverse step of Equation 1.4, the
addition of an amount
of heat sufficient to destabilize the adsorbed vapour will cause
water to be driven off the
adsorbent, regenerating the material to its maximum potential
for adsorption.
1.2.4 Thermochemical Materials
Salt Hydrates
Thermochemical heat is released in the hydration of anhydrous
salts, and stored by
dehydrating them with low grade heat (100◦C < T < 260◦C).
These salts can be hydrated
until they are essentially saturated during discharge, and can
then be dehydrated feasibly
to their monohydrate. In general, salt hydrates have relatively
high thermochemical
energy densities (300 - 400 kWh/m3), but lower discharge power
as shown by van Essen
et al [12]. During hydration, salt hydrates can also form a
gel-like structure which can
become a non-porous layer that limits mass transfer if dried as
researched by Zondag et
al. [13]. For some salt hydrates, corrosive gases are the
product of regeneration reactions
and material degradation may occur above a threshold
temperature.
6
-
Adsorbents
Solid adsorbent materials behave like desiccants and have an
affinity for water
vapour in the atmosphere, and adsorb and desorb gases according
to the three step
desiccant cycle of Figure 1.2.
350°C
1
2
20°C
3
ADSORBENT MOISTURE CONTENT ADSORBENT MOISTURE CONTENT
AD
SO
RB
EN
T S
UR
FA
CE
WA
TE
R
VA
PO
R P
RE
SS
UR
EA
DS
OR
BE
NT
SU
RF
AC
E W
AT
ER
V
AP
OR
PR
ES
SU
RE
AD
SO
RB
EN
T S
UR
FA
CE
WA
TE
R
VA
PO
R P
RE
SS
UR
EA
DS
OR
BE
NT
SU
RF
AC
E W
AT
ER
V
AP
OR
PR
ES
SU
RE
Figure 1.2: Adsorbent water vapour pressure as a function of
moisture con-tent (left) and the adsorption-regeneration cycle
(right), adapted from [14].
Similar to the saturation of salt hydrates, adsorbents have a
maximum capacity
for adsorption based on the vapour pressure of the adsorbate gas
(y-axis of Figure 1.2)
and the pressure at the surface of the adsorbent (abscissa of
Figure 1.2) as a function of
the moisture accumulated. The adsorption step from 1 - 2 results
in an adsorbent that
is at a higher temperature and a higher vapour pressure at its
surface that comes to
equilibrium with the surrounding air. Increasing the temperature
of the adsorbent from
steps 2 - 3 makes the vapour pressure of the water on it surface
to become greater than
that of the surrounding air, causing energized molecules of
water vapour to return to
the free phase (regeneration). Before the adsorbent can adsorb
again, it must cool from
steps 3 - 1 to decrease the vapour pressure at its surface.
Adsorbents such as zeolites
exhibit lower energy storage potential than salt hydrates, with
energy densities of 100 -
200 kWh/m3; however, they offer a high thermal stability and
rate of discharge [15].
7
-
Zeolites
Zeolites are crystalline aluminosilicates of alkali or alkali
earth elements such as
sodium, potassium, and calcium, and are synthesized and used in
industry for applica-
tions of water treatment and removing CO2 from flue gases [16].
Conventionally these
materials are shaped by manufacturers as beads or pellets,
containing up to 20 wt%
binder that is inert for adsorption [17].
Overall, Figure 1.3 shows how the energy densities of each mode
as a process alone
compare to the effective energy storage densities of their
systems which include the
added volume of other components that are specific to each
technology.
SensibleLatentAdsorptionChemical
Syst
em E
nerg
y D
ensi
ty (k
Wh/
m3 )
0
50
100
150
200
250
Mode Energy Density (kWh/m3)0 100 200 300 400 500
Figure 1.3: Thermal energy density of systems compared to mode,
adaptedby [18].
Despite offering a higher energy density as a mode of thermal
energy storage, it
can be seen that adsorption-based TES sytems are comparable to
sensible and latent
thermal energy stores as packaged systems. Because adsorption
TES systems are so
much earlier in development with most of the cited values for
energy density based
on research conducted using fixed adsorbent beds, this present
work was compelled to
investigate other potential contacting methods for this
application in pursuit of reducing
the size of the adsorber component.
8
-
1.3 Research Objectives
The main objective of this research was to obtain experimental
data on the ther-
modynamic adsorption behaviour of a 60x65 mesh zeolite 13X
molecular sieve under
fluidization and provide analysis regarding a scale up study
based on the results. The
following objectives were set to obtain this analysis,
including:
• conducting a comprehensive literature review of the state of
the art in compact
thermal energy storage;
• the design, construction, and commissioning of the
experimental setup, including
the plumbing, instrumentation, and the machining of an
adsorption column; and
• developing a procedure to evaluate fixed and fluidized beds as
components in TES
systems under various conditions of the adsorbate (flow rate,
concentration, pres-
sure, and temperature)
1.4 Contribution of Research
This work included the:
1. Development of an apparatus for comparing the adsorption
behaviour of fixed and
fluidized beds of zeolite 13X;
2. Experimental evaluation of the performance of the adsorbent
beds under various
charge and discharge conditions; and
3. Proposal of a methodology for scaling adsorption-based
thermal energy storages.
9
-
1.5 Organization of Research
This thesis is divided into the following chapters:
Chapter 2 presents a review of the current literature on
thermochemical energy
storage, including a comparison of the laboratory and
pilot-scale
studies that influenced this work;
Chapter 3 presents the physical design of the apparatus and an
outline of the
test procedure followed for this work;
Chapter 4 presents the results of the experimental study based
on temperature
and humidity profiles; and
Chapter 5 presents conclusions drawn from this study, and
discusses
recommendations for future work.
Appendices A to D present additional information that support
this research.
10
-
Chapter 2
Literature Review
This chapter contains a review of the literature on past and
current research activ-
ities focused on the potential of thermochemical materials to
provide thermal energy
storage for residential heating applications. The status of
thermochemical energy stor-
age within the landscape of international energy policy is
summarized, including the
major achievements of researching members of a task force on the
subject; a review of
thermochemical energy storage system types is given, with
emphasis on fluidization and
fluidized bed technology; and, an outline of the different
performance metrics used to
evaluate the performance of thermochemical energy storage
systems is then presented,
providing context for a series of comparative studies and the
experimental evaluation of
the system developed for this work.
2.1 Global Status of the Technology
2.1.1 The International Energy Agency
The International Energy Agency (IEA) is an autonomous body
within the Orga-
nization for Economic Co-operation and Development (OECD) that
has led the global
dialogue on energy since 1974. The IEA advocates policies that
will enhance the acces-
sibility, reliability, and sustainability of energy in its 29
member countries and beyond.
It achieves this work through Technology Collaboration Programs
(TCP) that provide
data and authoritative analysis on the full spectrum of energy
issues.
11
-
2.1.2 IEA - Solar Heating and Cooling Program
Established in 1977, the Solar Heating and Cooling Program (SHC)
is a TCP of the
IEA that promotes all aspects of solar thermal energy through
the collaborative efforts
of international expertise. Objectives of this program include
the research, development,
and demonstration of new solar thermal technologies. This work
is carried out through
a series of Tasks or research projects that study various
aspects of solar heating and
cooling. Each Task is managed by an operating agent from one of
the member countries
or sponsoring organizations, and overall control of the program
lies with an executive
committee.
2.1.3 IEA - SHC Task 42
Recognizing energy storage as an integral component of
energy-efficient systems, the
Energy Conservation and Energy Storage (ECES) Program of the IEA
is comprised of
a strategic plan for the research, development, dissemination,
and market deployment
of energy storage technologies. Task 42 is a joint Task with
Annex 29 of the ECES
titled Compact Thermal Energy Storage: Material Development for
System Integration,
which has an overall objective of developing advanced materials
and systems for the
compact storage of thermal energy. The objectives of Task
42/Annex 29 as outlined by
its position paper [19] were to:
• identify material requirements via numerical simulation of
known storage technolo-
gies;
• identify, design, and develop new and composite materials;
• develop reliable and reproducible measuring and testing
procedures to characterize
new storage materials;
• enhance the performance, stability, and cost-effectiveness of
new storage materials;
• develop multi-scale numerical models that describe and predict
the performance
of new materials;
• develop and demonstrate novel compact thermal energy storage
systems employing
the advanced materials; and
12
-
• develop a method for the economic evaluation of compact
thermal energy storage
systems
Activities for the Task were divided between five working groups
(WG) organized in
a matrix according to Figure 2.1.
Eco
no
mic
Eval
uat
ion
Ap
plic
ati
on
s an
d
Syst
em
In
tegr
ati
on
Material Engineering and Processing
Testing and Characterization
Numerical Modelling
WG_A1:
WG_A2:
WG_A3:
WG_B WG_C
Eco
no
mic
Eval
uat
ion
Ap
plic
ati
on
s an
d
Syst
em
In
tegr
ati
on
Material Engineering and Processing
Testing and Characterization
Numerical Modelling
WG_A1:
WG_A2:
WG_A3:
WG_B WG_C
Figure 2.1: Organization of IEA-SHC Task 42/ECES Annex 29,
adaptedfrom [19].
The WG forming the blue rows of Figure 2.1 represented three
categories related
to materials (A1, A2, and A3). The activities of these WG
intersected with Working
Groups B and C, which focused on system development and the
economic evaluation
of compact thermal energy storage systems, respectively. The
Task officially started
January 1, 2009 and ended December 31, 2015.
The key achievements of each WG for Task 42/Annex 29 pertaining
to thermochem-
ical storage are summarized in Tables 2.1 - 2.5.
13
-
Table 2.1: Key achievements by Task 42/Annex 29 Working Group
A1.
Member of WG A1 Activities
Leuphana University(Germany)
• using a reactor simulation, found SrBr2·6H2O(392 kWh/m3),
LaCl3·7H2O (359 kWh/m3),and MgSO4·6H2O (340 kWh/m3) as
potentialcandidates out of a study of 125 salt hydratesbased on
thermodynamics and non-toxicity [20]
• experiments limiting water vapour flowdemonstrated the
possibility to avoid theformation of solutions (improving cycle
stability)with hydration; however, this was found to beinfeasible
at the macro-scale [21]
• the impregnation of salt hydrates into vermiculiteimproved
cycle stability and thermal conductivitybut decreased energy
density [22]
Universities of Lleida& Barcelona (Spain)
• from corrosion tests between inorganic TCMand metal reactor
vessels, recommend that only316 stainless steel should be used for
reactorscontaining CaCl2, Na2S, CaO, MgSO4,and MgCl2 [23]
• the thermo-physical characterization of CaCl2(408 kWh/m3) and
zeolite (55 kWh/m3)was examined [24]
TH-Wildau (Germany)
• modified the hydrophilic character of NaYzeolites via
dealumination through a steamingprocess, shifting their adsorption
isothermto permit regeneration at temperatures lessthan 100◦C
[25]
14
-
Table 2.2: Key achievements by Task 42/Annex 29 Working Group
A2.
Member of WG A2 Activities
TH-Wildau (Germany)
• characterized two binder-free zeolites:13XBF and NaYBF
(Chemiewerke Bad Köstritz)which showed water adsorption capacities
slightlygreater than 30 wt.%, with no degradation undertemperatures
of 180 to 200◦C [26]
Eindhoven University ofTechnology (The Netherlands)
• characterized the dehydration of two referencematerials
(Li2SO4·H2O and CuSO4·5H2O) byTG-DSC thermal analyses, and in-situ
XRDmeasurements and microscopic observations onmonocrystals
[27]
• compared to sulfates such as MgSO4·7H2O(589 kWh/m3), chlorides
such as MgCl2·6H2O(508 kWh/m3) were found to offer a lowerenergy
density, but a higher temperature liftand better stability if
restricted to a 40 - 130◦Coperating range [27]
Table 2.3: Key achievements by Task 42/Annex 29 Working Group
A3.
Member of WG A3 Activities
Gaastra-Nedea et al. [28]
• simulations of molecular dynamics (MD) of a MgCl2system showed
a similar trend to that found fromthermodynamic equilibrium: a
higher water vapourpressure counteracts both the dehydration
andhydrolysis reaction whereas a higher temperatureincreases the
amount of dehydration and hydrolysis
• hydrolysis occurring from the di-hydrate was predictedto be
unlikely under the conditions of the heat storagesystem, while MD
simulations showed that up until awater vapour pressure of 20 - 40
mbar, HCl is formed
• MD results for three different water vapour pressuresshowed
that a higher vapour pressure ensures a morecrystalline stucture
with a lower diffusivity of water
15
-
Table 2.4: Key achievements by Task 42/Annex 29 Working Group
B.
Member of WG B Activities
ITW University ofStuttgart (Germany) [29]
FlowTCS Project:
• developed an open sorption process with zeolite andsalt
impregnated zeolite in an external, quasi-continuous,cross-flow
reactor concept of 30 L volume coupled with200 L of material
storage
CETHIL [30]
STAID Project:
• developed an 80 kg zeolite system consisting of two40 kg
sub-reactors that could be configured in seriesand in parallel; the
serial configuration developed amaximum power of 2.25 kW
ZAE Bayern [31]
Mobile Sorption Heat Storage:
• developed a zeolite sorption storage mounted on asemi-trailer
that was charged using waste heat froman incineration plant
(130◦C), and then driven to aindustrial drying process where heat
from the dischargeof the storage (160◦C) was used to reduce the
usenatural gas
• the storage contained 14 tons of zeolite in a 22 m3volume and
could realize a capacity of 2.3 MWh withadsorption inlet conditions
of 60◦C and 0.009 kgH2O/kgda
16
-
Table 2.5: Key achievements by Task 42/Annex 29 Working Group
C.
Member of WG C Activities
Rathgeber et al. [32]
• developed and tested a tool for the economic evaluationof 26
existing thermal energy storage projects for specificusers and
applications
• determined that the annual number of storage cycleshad the
largest influence on the cost effectiveness of thestorages, and
that a major fraction of the investmentcosts for the investigated
projects were not of thestorage material itself but costs of the
storagecontainer or reactor including the chargingand discharging
unit
2.2 Systems for Thermochemical Storage
2.2.1 Open vs. Closed Systems
A thermochemical process employing a solid/water pair can
operate according to
two different modes: the solid can react with pure water vapour
at low pressure in a
closed system, or with a moist air flow at atmospheric pressure
in an open system (see
Figure 2.2).
Domestic Hot Water
Solar Thermal/Waste Heat
Inlet Air (Fan or Compressor)
Outlet Air (Space Heat or Ambient)
Water Vapour(Borehole or Humidifier)
Charge
DischargeDischarge
Figure 2.2: Open-system concept for thermochemical energy
storage.
17
-
In Figure 2.2, the adsorber refers to the component of the
open-system in which
the adsorption and regeneration processes take place. During
adsorption (discharge),
air driven by a fan or a compressor is mixed with water vapour
from a humidifier or
a borehole heat exchanger to produce the water vapour adsorbate.
The heat released
in the adsorber can be transferred to an air-to-water heat
exchanger for heating DHW,
or can be supplied directly to heating a space. During
regeneration (charge), dry air
entering the adsorber could be heated by a loop circulating an
appropriate heat transfer
fluid for solar thermal energy or low-grade waste heat from an
industrial process. The
moist air leaving the adsorber would be released to the ambient
environment, instead of
being condensed and re-used in the case of a closed system.
Integrated vs. External Systems
Two main concepts for adsorbers are described in the literature:
the integrated
type in which adsorption occurs within the entire storage of the
material itself, or the
external type where the active adsorbent material is kept
separate from the adsorber
(see Figure 2.3).
material segment
heat exchanger
moi
st a
ir
adsorber
heat exchanger
spent material(saturated)
dry material(regenerated)
material storage
solids in
solids out
moist air
Figure 2.3: Integrated (left) vs. external (right) adsorption
systems, adaptedfrom [33].
18
-
The integrated concept is technologically easier to realize as
it requires no additional
mechanisms and input energy for material transport as the
adsorbent is stationary within
the storage. During charge of the system however, studies have
shown that higher
regeneration temperatures are required, implying that
temperature resistant materials
are required throughout the entire storage system [33]. Because
the material storage and
adsorber are the same component, this concept is better suited
for diurnal applications
where the system is charged and discharged as single cycles.
The external concept is technologically more complex as it
requires material transport
between the material storage and the adsorber. The adsorbent
material used must also
be abrasion resistant to be transportable. Adsorption and
regeneration are however
reduced to only a fraction of the total material at a time. This
gives the external
concept the advantage of reduced heat losses during
regeneration, and gives the system
more flexibility in terms of the rate at which material is
charged and discharged
2.3 Fundamentals of Fluidization
As shown by Figure 2.4, fluidization is a phenomenon in which a
bed of fine solids
exhibits fluid-like behaviour through contact with an upward
moving gas or liquid, i.e.,
a fluidized bed of solids will not resist an applied shear or
tangential stress by a static
deflection, but move and deform continuously as long as the
stress is applied [34].
Figure 2.4: Liquid-like behaviour of gas fluidized beds, adapted
from [35].
19
-
The fluidized solids in the bed of Figure 2.4 respond to the
applied stress of an
immersed object by moving around it completely in the case of
the sinking heavier object;
lighter objects pushed into the bed will pop up and remain
floating at the surface [35].
This contrasts the expected behaviour from a stagnant
non-fluidized bed of solids which
would respond to the weight of the object with an equal and
opposite force prohibiting
it from sinking into the bed. Should the fluidized bed be
tipped, the solids will respond
by maintaining a horizontal surface, and if a hole were to be
made in the side of the
bed, the solids would spew from the hole in the form of a
jet.
The degree of contact observed between a batch of solids and a
fluid depends on the
balance of the force of gravity (weight) and force of drag
acting on individual particles.
These forces depend on physical properties including the
diameter of the particle (dp)
and the difference in the density between the fluid and the
particle (see Figure 2.5).
Figure 2.5: Balance of forces acting on fluidized solid
particle.
Referring to the fluidization regimes of Figure 2.6, at a low
flow rate, the fluid will
simply pass through the void spaces of stationary particles in
what is referred to as a
fixed bed. All regimes of fluidization therefore begin with a
fixed bed in which the force
of gravity is greater than that of the force of drag on
individual particles of the bed. At
higher gas velocities, the force of drag will increase
proportionally with the flow rate of
the fluid to counterbalance the force of gravity, suspending the
particles in a state of
minimum fluidization.
20
-
Figure 2.6: Fluidization regimes of gas fluidized beds, adapted
from [35].
Increasing the flow rate beyond minimum fluidization for
gas-solid systems, however,
will result in large instabilities producing the bubbling and
channeling of gas as shown
for the bubbling fluidization regime of Figure 2.6. The vigorous
agitation of the solids
at higher velocities becomes more violent, providing a high
level of mixing in a bed that
does not expand significantly beyond its volume observed at
minimum fluidization [35].
While the quality of fluidization in a gas-solid bed will be
largely determined by the
properties of the solid and fluid alone (see Section 1.5.1),
factors such as gas flow rate,
bed geometry, and internal components of the bed will influence
the rate of solid mixing
in the bed. Referring to Figure 2.6, slugging is a phenomenon
that is occurs in long,
narrow beds that allow bubbles to grow to the point that they
become large enough
to push a slug of cohesive solids upwards. The slug eventually
disintegrates as another
slug forms and this behaviour continues in oscillation. Such
behaviour is undesirable
for industrial applications due to the lack of solid-gas
contacting present in this regime.
Lastly, lean phase fluidization or pneumatic transport will
occur at the highest veloci-
ties at which solids are carried out of the bed, providing
little opportunity for contact
between the solid and fluid.
21
-
Particle Size and Fluidization Velocity
The quality of fluidization relating to the size and coalescence
of bubbles in a bubbling
bed largely depends on the diameter of the particles (d p) and
the difference between the
density of the solid and gas phases (ρs– ρg). While fluidization
quality also depends
on the geometry of the bed, non-ideal behavior such as slugging
occurs when cohesive
forces between the solids are greater than forces of drag.
Geldart [36] classified particles according to these properties
into four groups: A, B,
C, and D, which display different fluidization characteristics
based on the cohesion forces
between particles for each group. Compared to group B particles
that begin bubbling
immediately at the minimum fluidization velocity (umf), a bed of
group A particles will
exhibit unique fluidization behavior, by expanding in height
over the minimum fluidiza-
tion velocity before bubbling at a higher velocity. Enhanced
gas-solid contact efficiency
has been observed with this behavior, making group A particles
ideal for industrial ap-
plications [37]. Additionally, Shaul et al. [38] have observed
that the pressure drop in
fluidized beds of group A particles is not affected by the
height to diameter (H/D) ratio
of the bed, placing less technological constraints on design and
scale-up. The pressure
drop within a bed of group B solids however will increase with
the H/D ratio of material.
Group C particles are so fine and cohesive that they do not
exhibit bubbling fluidization
but channel and slug according to Figure 2.6. Group D particles
exhibit a less efficient
spouting fluidization at significantly higher gas velocities
than other groups due to the
higher force of drag required to act over a larger particle
diameter.
This present work is concerned with smaller particles from group
A which will provide
the fluidized case and larger particles from group D which will
provide the fixed bed case.
Geldart [36] defined group A of solids as having a small mean
size and/or a density of less
than 1.4 g/cm3. With a bulk density of 0.64 g/cm3, zeolite 13X
particles in the range of
50 to 300 µm should exhibit group A fluidization characteristics
according to Geldart’s
empirical powder classification diagram. A 60x65 Tyler mesh, or
sieved zeolite of 210
- 250 µm in particle diameter was prepared for this research
based on the assumption
that it would behave as a group A solid.
22
-
Per Kunii and Levenspiel [35], the minimum fluidization velocity
of smaller particles
belonging to this group can be calculated according to:
umf =dp
2(ρs − ρg)g1650µ
Rep < 20 (2.1)
For comparison, the minimum fluidization velocity of large
particles can be calculated
according to:
umf =
[dp(ρs − ρg)g
24.5ρg
] 12
Rep > 1000 (2.2)
For 0.001 < Rep < 4000, Equation 2.1 and have been found
to give predictions of umf
with a standard deviation of ±34% [39]. To avoid carry-over of
solids from a bed, the
gas velocity should be kept somewhere between the umf of the
mean particle diameter
and the terminal velocity (ut) of the smallest diameter present
for the size distribu-
tion in appreciable quantity. The terminal velocity can be
obtained from experimental
correlations of the dimensionless groups CdRep2 versus Rep,
where:
Rep =dpρgutµ
(2.3)
and the velocity independent group is defined as:
CdRep2 =
4gdp3ρg(ρs − ρg)
3µ2(2.4)
2.4 Performance Metrics
Thermochemical energy storage systems can be compared using a
number of different
parameters to evaluate the performance of thermochemical
materials and systems as a
whole. Some of the key performance parameters used within the
literature and this work
are discussed in the following paragraphs.
23
-
Energy Released
The energy released (discharged) is a measurement of the amount
of heat discharged
over the length of hydration or adsorption, and is calculated by
measuring the change
in the thermodynamic properties of the air into and out of the
TES system according
to:
Qd = ρV̇ cp∆T (tn − tn−1) (2.5)
For Equation 2.5, the mass term of Equation 1.1 is replaced by
the mass flow rate
of the water vapour taken up by the TCM over the length of
hydration or adsorption
per time step (tn – tn−1), or 10 seconds for the data
acquisition (DAQ) system used in
this work. The density and specific heat capacity of Equation
2.5 are of moist air, and
∆T is the temperature lift, or the difference in temperature at
the inlet and outlet of
the system.
Energy Density
The energy density is an indication of the compactness of a TES
system and is most
commonly expressed in terms of kWh/m3. Studies in the literature
typically define
energy density in terms of the material and the prototype as a
whole considering their
respective volumes:
Energy Density =Energy Released
Column or Material Volume(2.6)
Specific Power and Storage Capacity
The specific power of a TES is the average rate at which heat
was released per mass
of TCM during hydration or adsorption [W/kg]. The storage
capacity considers the
specific power of the system released over the length of the
hydration or adsorption
process [Wh/kg].
Storage Capacity or Specific Power =Energy or Rate of Heat
Released
Mass of Material(2.7)
24
-
2.5 Comparative Studies
2.5.1 Material Characterization and Selection
A high hydration or adsorption capacity (water uptake) is
essential for a high energy
density, and high kinetics are required for a high thermal power
output during the
discharge of a thermochemical system. TCM candidates must also
offer high cyclability
(i.e., not degrade over time), and for the built environment,
they must be non-toxic.
Salt Hydrates
Considering these requirements, N’Tsoukpoe et al. (2014)
conducted a three-step
screening process for identifying suitable material candidates
for low temperature ther-
mochemical energy storage [20]. Out of 125 material
candidates:
• 45 salts passed an initial discrimination based on toxicity
and flammability;
• 17 salts showed reversibility from TGA measurements; and
• 3 salts (SrBr2·6H2O, LaCl3·7H2O, and MgSO4·6H2O) were selected
for further
study
The three most promising candidates, however, tended to form
partial solutions during
hydration, which led to the formation of non-porous layers,
limiting vapour diffusion
during dehydration processes. In addition to these limitations,
it was determined that
these materials don’t offer the energy density or heat recovery
efficiency to compete with
simpler technologies such as sensible heat storage.
van Essen et al. [12] from the Energy Research Centre of The
Netherlands (ECN)
studied the hydration and dehydration behaviour of MgSO4 through
thermogravimetric
analysis (TGA) and differential scanning calorimetry (DSC). From
these experiments,
it was observed that 88% of stored energy was released from
samples exposed to air at
25◦C with a water vapour pressure of 2.3 kPa. While MgSO4 was
shown to store almost
nine times more energy than can be stored in water as sensible
heat, only a maximum
temperature lift of 4◦C was observed.
Researchers at ECN also investigated the application of
MgCl2·6H2O for seasonal
heat storage [13]. Investigations by TGA and an evacuated
reactor setup showed that
25
-
MgCl2·6H2O started to form a mono-hydrate at 120◦C, and began
forming corrosive HCl
at temperatures above 135◦C. During hydration, material was
found to form a gel-like
structure that increased the pressure drop across the bed in a
larger reactor.
Lele et al. (2015) measured the thermal conductivity of material
candidates for ther-
mochemical storage using DSC [40]. Thermal conductivities of 0.3
- 1.3 W/m·K were
observed obtained from the four salts studied. The use of a
carrier matrix was discussed
as a potential method of improving conductivity and mitigating
the formation of gels
during hydration.
Adsorbents
Through numerical modeling validated by experimental work, Mette
et al. (2014)
have shown that zeolite 13X is characterized by a high water
uptake and fast kinetics [41].
The experimental results show the influence of the wall of a
fixed bed reactor on the
temperature lift profile, with a higher bed porosity in near
wall regions having led to
faster completion of adsorption.
Cortés et al. (2010) used TGA and mass spectrometry to study
the equilibrium
adsorption of water vapour on zeolite 13X [42]. Similar to the
calorimetric measurements
of Mette et al. (2014), it was observed that as the uptake of
water increased from 0.06
- 0.22 gH2O/g13X , the energy released decreased from 3,430 to
3,010 J/gH2O. From this
result, it was inferred that when the zeolite is close to its
maximum uptake at equilibrium,
the thermodynamic properties of adsorption are similar to those
of pure water.
For the application of separating CO2 from flue gases, Hauchhum
and Mahanta in-
vestigated the adsorption behaviour of a variety of zeolites
including 13X in a fixed
bed [43]. These investigations were extended to a fluidized bed,
in which the adsorption
capacity of zeolite increased by 1.9 molCO2kg−113X [44].
Details of the experimental setup and results for the foregoing
characterization of
adsorbents are summarized in Table 2.6.
26
-
Table 2.6: Characterization of adsorbents for thermochemical
storage.
Author(s) Experimental Setup Performance
Mette et al. [41]
Material:170 g of 8x10 mesh,binderless zeolite 13X
Sorbate: H2O gas(PH2O: 5 - 30 mbar)
Apparatus: fixed bed,H=140 mm, ID= 50 mm(H:ID of 2.8)
Inlet Temperature(s):30 & 50◦C (adsorption)180◦C
(desorption)
Temperature Lift:
10◦C @ 5 mbar (Tin= 30◦C)
37.5 ◦C @ 15 mbar (Tin= 30◦C)
35◦C @ 15 mbar (Tin= 50◦C)
75◦C @ 30 mbar (Tin= 30◦C)
Hauchhumand Mahanta [43]
Material:zeolite 13X (20 g),8x12 mesh
Sorbate: CO2 gas(PCO2 : 0 - 1 bar)
Apparatus:packed-fixed bed200 mm long, 27 mm ID(H:ID of 7.4)
Flow Rate:15 L/min (adsorption)
Inlet Temperature(s):25 - 60◦C (adsorption)
CO2 Adsorbed:4.215 molCO2kg
−113X
27
-
Table 2.6 - continuedAuthor(s) Experimental Setup
Performance
Hauchhumand Mahanta [44]
Material:zeolite 13X (200 - 300 g),8x12 mesh
Sorbate: CO2 gas(PCO2 : 0 - 4 bar)
Apparatus: fluidized bed1300 mm long, 50 mm ID,(H:ID of 3.2 -
4.7),310 - 462 mL
Flow Rate: 15 L/min
Inlet Temperature(s):25 - 60◦C (adsorption)
CO2 Adsorbed:6.13 molCO2kg
−113X
2.5.2 Lab-Scale Prototypes
Open Lab-Scale Systems
Dicaire and Tezel from the University of Ottawa investigated the
effects of adsorption
flow rate, regeneration temperature, feed air humidity, and
cycling on a hybrid adsorbent
comprised of activated alumina and zeolite 13X [45]. Adsorption
flow rate and cycling
were shown to have no effect on the energy density of the AA13X
sample; however,
increasing the regeneration temperature by 170◦C was found to
increase energy density
by 50 kWh/m3. Using the same experimental setup, Lefebvre et al.
[46] conducted a
parametric study which examined the effect of column dimension,
particle diameter,
adsorption activation energy, flow rate, column void fraction,
and adsorbent heat of
adsorption on the performance of an adsorption-based system.
Nonnen et al. (2016) demonstrated long-term heat storage based
on the sorption
of water on a CaCl2 and Na-X composite material in both a
lab-scale apparatus and
hardware-in-the-loop (HIL) test bench [47]. The work concerning
the HIL test-bench is
unique in the literature for considering the concepts of a
pre-reactor and gravity driven
material transport.
28
-
Zondag et al. (2013) at ECN developed a prototype capable of
providing 50 W of
thermal power to a load for 40 hours through the hydration of
MgCl2·2H2O, indicating
an effective energy density of approximately 139 kWh/m3 [48].
From the pressure drop
measured across the reactor bed, the prototype was determined to
have a COP of 12.
Details of the experimental setup and performance of each open
lab-scale prototype
are summarized in Table 2.7.
Table 2.7: Lab-scale testing of open-systems.
Author(s) Experimental Setup Performance
Dicaireand Tezel [45]
Material:55 g of 8x12 mesh activatedalumina and zeolite
13X(AA13X)
Sorbate: H2O gas(RH: 100% for adsorption,0 & 50 % for
desorption)
Apparatus: packed-fixed bed70 mm long, 38.1 mm OD(62.76 mL, H/D
Ratio of 1.8)
Flow Rate:24 L/min (adsorption),8 - 24 L/min (desorption)
Inlet Temperature(s):20 ◦C (adsorption),80 - 250◦C
(desorption)
Temperature Lift: 55◦C
Energy Density:197 kWh/m3 ± 7 kWh/m3
Regeneration:2 - 3 hours
Cycling:50 runs with no effect onenergy density
Storage Efficiency:30 - 60 % for 80 - 250◦C
29
-
Table 2.7 - continuedAuthor(s) Experimental Setup
Performance
Nonnen et al. [47]
Material: 15CaCl2/Na-X
Sorbate: H2O gas(PH2O: 15 mbar )
Apparatus:lab-scale apparatus :packed-fixed bed,7 cm3 bed
volume
hardware-in-the-loop (HIL)test bench: gravity driven,24 L
material volume
Flow Rate:0.3 kg/h (lab-scale),150 kg/h (HIL)
Inlet Temperature(s):
Discharge:30◦C lab-scale, 28◦C HIL
Charge:110◦C lab-scale, 180◦C HIL
Temperature Lift:35◦C in HIL
Discharge Power:1.2 kW (HIL test bench)
Energy Density:260 kWh/m3 (lab-scale)
Zondag et al. [48]
Material: MgCl2·6H2O (17 L)
Sorbate: H2O gas(PH2O: 12 mbar )
Apparatus: packed bed
Flow Rate: 510 L/min
Inlet Temperature(s):
Hydration:25◦C (Tevap =10
◦C)50◦C (Tevap =25
◦C)
Dehydration: 130◦C
Temperature Lift:62◦C @ Tin = 25
◦C18◦C @ Tin = 50
◦C
Discharge Power:50 - 150 W
Energy Density:139 kWh/m3
Pressure Drop:100 Pa (packed bed),220 Pa (system)
30
-
Closed Lab-Scale Systems
With the goal of developing a large-scale thermochemical energy
storage to be in-
tegrated with industrial processes and HVAC systems, Lass-Seyoum
et al. (2012) con-
ducted adsorption experiments for various zeolites at the
lab-scale in 1.5 L and 15 L
closed-system apparatus. Suitable storage materials and process
conditions were char-
acterized in the 1.5 L system, while heat exchanger concepts
were optimized in the
development of the 15 L system [49].
Details of the experimental setup and performance of each closed
lab-scale prototype
are summarized in Table 2.8.
Table 2.8: Lab-scale closed-system thermochemical projects.
Author(s) Experimental Setup Performance
Lass-Seyoum et al. [49]
Material: various zeolites(800 - 900 g), 8x12 mesh
Sorbate: H2O gas(PH2O: 42 mbar )
Apparatus: packed bed (1.5 L)
Inlet Temperature(s):30◦C (adsorption),90 - 200◦C
(desorption)
Storage Capacity:160 - 220 Wh/kg
Specific Power:60 - 240 W/kg
Lass-Seyoum et al. [49]
Material: ∼15 L of variouszeolites, 8x12 mesh
Sorbate: H2O gas
Apparatus: packed bed (15 L)
Inlet Temperature(s):20◦C (adsorption),90 - 200◦C
(desorption)
Storage Capacity:180 - 240 Wh/kg
Specific Power:45 - 66 W/kg
31
-
2.5.3 Pilot-Scale Prototypes
Open Pilot-Scale Systems
From a 2D analysis of the second-law applied to a model of an
open and closed
system [50], Michel et al. developed a large-scale prototype
involving the hydration
of 400 kg of SrBr2 in an open system [51]. The mass flow rate of
the moist air was
determined to have a strong influence on the reaction kinetics
and thermal power of
the reactor; increasing the flow rate of the moist air from 150
to 313 m3/h resulted in
a 500 W increase in thermal power released during hydration.
Based on specifications
requiring a supply of 2 kW of heat over a 2 hour period,
Johannes et al. (2015) developed
a high-powered open storage system comprised of two 40 kg beds
of zeolite 13X that
were configured in series and parallel [30]. From investigations
of two regeneration
temperatures (120 & 180◦C), two flow rates (120 & 180
m3/h), and two levels of relative
humidity (50 & 70%), it was found that:
1. A lower regeneration temperature results in a lower amount of
energy released,
but at the same maximum power.
2. A higher humidity during adsorption results in a higher
maximum released power.
3. A reduction in the air flow rate leads to a reduction of the
maximum released
power with the same amount of stored energy released.
Gaeini et al. developed a 250 L open storage system capable of
discharging 3.6 kW
of heat with its four reactor segments running in parallel; each
segment could provide up
to 0.9 kW for 10 hours [52]. Over an 18 hour adsorption period,
each segment released
an average of 12 kWh of heat which equates to almost 70 % of the
heat supplied to each
segment during the 7 hour regeneration process. By reducing
non-reactive zones within
the system, Gaeini et al. expect to increase the energy density
of the setup to almost
145 kWh/m3.
Details of the experimental setup and performance of each open
pilot-scale prototype
are summarized in Table 2.9.
32
-
Table 2.9: Open pilot-scale thermochemical projects.
Author(s) Experimental Setup Performance
Michel et al. [51]
Material:400 kg of SrBr2·6H2O
Sorbate: H2O gas(4.3 - 6.2 gH2O/kgda @ 25
◦C)
Apparatus:modular porous fixed bed,8 rectangular modules(8x50 kg
of hydrated salt),L=74.9 cmW=72 cmH=96.6 cm,(0.52 m3 or 520 L)
Flow Rate:290 3/h (hydration),313 m3/h (dehydration)
Inlet Temperature:25 ◦C (hydration),80 ◦C (dehydration)
Discharge Power:300 - 800 W
Specific Power:0.75 - 2 W/kg
Energy Density:203 kWh/m3 (reactor),388 kWh/m3 (salt bed)
Storage Capacity:105 kWh (262.5 Wh/kg)
Pressure Drop:60 to 560 Pa (bed)
Johannes et al. [30]
Material:80 kg of Zeolite 13X
Sorbate:H2O gas 50 & 70% RH
Apparatus:two packed beds(40 kg of zeolite each)
Flow Rate:120 & 180 m3/h
Inlet Temperature:20 ◦C (adsorption),120 & 180◦C
(desorption)
Temperature Lift:37◦C
Discharge Power:1,410 - 2,249 W
Specific Power:27.5 W/kg
COP: 1.7 - 6.8
Storage Efficiency:34 - 54.1 %
33
-
Table 2.9 - continuedAuthor(s) Experimental Setup
Performance
Gaeini et al. [52]
Material:170 kg of Zeolite 13XBF
Sorbate: H2O gas(5.2 gH2O/kgda)
Apparatus: packed bed,four cuboid 62.5 L segments(250 L), 2:1
(W:H) aspect ratio
Flow Rate:50 g/s (adsorption)33 g/s (desorption)
Inlet Temperature:10◦C (adsorption),190◦C (desorption)
Temperature Lift:18◦C
Discharge Power:3.6 kW for 10 hours
Storage Capacity:17 kWh (100 Wh/kg)
Energy Density:198 kWh/m3 (material)108 kWh/m3 (prototype)
Storage Efficiency:67 %
Pressure Drop:30 Pa (bed)
Closed Pilot-Scale Systems
Lass-Seyoum (2016) conducted experiments on a large-scale (750
L) adsorption ther-
mochemical store in combination with solar thermal heat pipes.
Adequate thermal
energy for hot water and space heating (i.e., temperatures above
60◦C) were obtained
even after succeeding discharge processes over 4 days [53].
Finck et al. (2013) designed an adsorption heat storage module
based on a desired
capacity of 3 kWh for supplying space heat with a temperature
lift of 20◦C at a rate of
800 W [54]. Decreasing the temperature difference between the
condenser and the inlet
for desorption from 100 to 80◦C resulted in a 30% reduction in
the energy released by
the module [55].
Details of the experimental setup and performance of each closed
pilot-scale proto-
type are summarized in Table 2.10.
34
-
Table 2.10: Closed pilot-scale thermochemical projects.
Author(s) Experimental Setup Performance
Lass-Seyoum et al. [53]
Material: 750 L of zeolite(type unknown)
Sorbate: H2O gas
Apparatus: 0.75 m3,stationary bed (Solarlux)
Flow Rate:2.5 m3/h “heat carrier”,1.8 m3/h “cooling cycle”
Inlet Temperature:25 - 30◦C (adsorption)100 - 120◦C
(desorption)
Energy Density:200 - 220 kWh/m3
Temperature Lift:40 ◦C
Storage Capacity:150 - 220 Wh/kg
Specific Power:19 - 50 W/kg
Finck et al. [54, 55]
Material: zeolite 5A(40 kg), 8x12 mesh
Sorbate: H2O gas
Apparatus: packed bed,eight 1000x300x33 mm blockscontaining
zeolite with acylindrical vessel
OD=502 mmH=1,454 mm
Inlet Temperature:20◦C (adsorption)103◦C (desorption)
Energy Density:47 kWh/m3 (material),12.5 kWh/m3 (overall)
Temperature Lift:33◦C
Storage Capacity:63 Wh/kg
Specific Power:24 W/kg, 740 - 971 Wtransferred to load
35
-
2.6 Research Direction
To date, the main focus of research projects on compact
thermochemical energy
storage has been on material characterization and development,
the bulk of which have
been conducted in Europe, with fewer research activities taking
place in North America.
A review of the literature on thermochemical energy storage has
highlighted that research
and development initiatives on the systems aspects of the
technology are required to
match efforts that have been made to characterize and develop
new materials.
Given its capacities in advanced building mechanical systems,
Canada has the poten-
tial to carry out the applied research and development to
commercialize the mechanical
systems of this technology. In their 2012 assessment of the
technology, SAIC Canada
concluded that key developments required to realize
thermochemical storage are reac-
tor & adsorber designs, methods of transporting charged and
discharged material, and
methods of determining the state of charge of the system [8].
Many designs and methods
exist for contacting the solid TCMs with air; however, most of
the bench and pilot-scale
systems reviewed were of the integral type based on packed-bed
designs because of their
design simplicity.
Despite being used extensively in industry for their noted
advantages of higher heat
and mass transfer compared to fixed beds, fluidized beds have
only been considered con-
ceptually for the application of low-temperature thermochemical
energy storage. Exper-
imental studies in other fields have only focused on the pure
adsorption characteristics
of zeolite in fluidized beds (i.e., breakthrough and capacity).
Therefore, there is a lack
of thermodynamic data to support the evaluation of these systems
for thermal energy
storage. The proposed research direction for this work was
therefore to investigate the
thermodynamic behaviour (i.e., temperature lift and energy
released) of zeolites in a
fluidized bed and compare their performance to a fixed bed of
equal size, drawing con-
clusions regarding the scaling of the two bed types.
36
-
Chapter 3
Experimental Design
This chapter provides a detailed description of the experimental
apparatus devel-
oped for evaluating fixed and fluidized beds of zeolite 13X for
the application of thermo-
chemical energy storage. Following this description,
methodologies regarding material
preparation, experimental procedure, and calculations are
described.
3.1 The Experimental Setup
The experimental setup that was designed, constructed, and
commissioned during
this study is shown as Figure 3.1.
Figure 3.1: The experimental setup including (a) the pressure
and flow con-trol (b) the in-line air heater and bubbler (c) the
adsorption column.
37
-
3.1.1 Adsorption Configuration
Figure 3.2 is a piping and instrumentation diagram (P&ID) of
the system config-
ured for the adsorption experiments performed in this work.
Compressed air from the
building’s supply was made available in the laboratory by means
of an industrial quick-
disconnect tap on the bench. This air was first passed through a
manual pressure
regulator that reduced the pressure of the supply air to 60 psig
and filtered out oil, wa-
ter, and other particles that could be present in the line. The
air was further regulated
by an I/P converter which could remotely regulate the pressure
of the air between 0 and
60 psig with a 4 - 20 mA signal.
ColumnColumn
Filter & Regulator
I/P Converter
Wet MFC
Bubbler
Air Heater
P1
P2
RH1
P3 RH2
Water Trap
Filter
Air Tap(Supply)
Dry MFC
Bypass
Bellmouth Inlet
Open Ball Valve (Permitting Flow)
Closed Ball Valve (Prohibitng Flow)
Open Ball Valve (Permitting Flow)
Closed Ball Valve (Prohibitng Flow)
Figure 3.2: The setup configured for adsorption experiments.
Before the air was split between a dry and wet pathway, a
pressure transducer (P1)
measured the pressure of the air entering the apparatus, which
was used to indicate the
amount of pressure drop across the system with additional
measurements made at P2
and P3. The humidity (mixing ratio) and therefore the vapour
pressure of the water
entering the adsorption column was controlled by two mass flow
controllers (wet and dry
MFC) with a 0 - 5 Vdc signal. The flow controlled by the dry MFC
was passed through
a 500 W in-line air heater controlled by potentiometer and a
feedback loop to maintain
38
-
a constant temperature of 25◦C at the inlet of column. The flow
controlled by the wet
MFC was bubbled through a vessel containing water; a water trap
following the bubbler
prevented liquid water from entering the adsorption column.
The relative humidity and temperature of the air entering and
exiting the adsorption
column were measured by two high-temperature relative humidity
transmitters (RH1
and RH2 respectively). Signals from two MFCs in conjunction with
the inlet relative
humidity measurement were used in a PID control loop to maintain
a set mixing ratio
at the inlet of the adsorption column (see subsection 3.1.6) .
The bypass allowed for the
mixing ratio and temperature of the air to stabilize prior to
introducing the air to the
column at the beginning of an experiment.
3.1.2 Regeneration Configuration
Non-Vacuum Regeneration
Figure 3.3 is a P&ID of the bench-scale adsorption TES
system configured for in-
situ regeneration of the adsorbent under the near atmospheric
pressure of the building’s
compressed air supply.
ColumnColumn
Filter & Regulator
I/P Converter
Wet MFC
Bubbler
Air Heater
P1
P2
RH1
P3 RH2
Water Trap
Filter
Air Tap(Supply)
Dry MFC
Bypass
Bellmouth Inlet
Open Ball Valve (Permitting Flow)
Closed Ball Valve (Prohibitng Flow)
Open Ball Valve (Permitting Flow)
Closed Ball Valve (Prohibitng Flow)
Figure 3.3: The setup configured for in-situ regeneration.
39
-
The configuration of Figure 3.3 is similar to that shown in
Figure 3.2, except that
the flow rate of the wet MFC is set to 0 L/min and the ball
valve following the water
trap is closed, directing the air exclusively through the dry
MFC pathway.
Partial Vacuum Regeneration
Figure 3.4 is a diagram of the system configured for in-situ
regeneration under a
partial vacuum. Similar to Figure 3.3, air is only able to flow
through the dry MFC
pathway; however, a material conveying vacuum pump (Vaccon DF
2-3) located at the
outlet of the system is used to pull air ambient air into the
apparatus at the bellmouth
inlet.
ColumnColumn
Filter & Regulator
I/P Converter
Wet MFC
Bubbler
Air Heater
P1
P2
RH1
P3 RH2
Water Trap
Filter
Air Tap(Supply)
Dry MFC
Bypass
Bellmouth Inlet
Open Ball Valve (Permitting Flow)
Closed Ball Valve (Prohibitng Flow)
Open Ball Valve (Permitting Flow)
Closed Ball Valve (Prohibitng Flow)
Vacuum Pump
Figure 3.4: The setup configured for in-situ vacuum
regeneration.
3.1.3 The Adsorption Column
The zeolite that underwent the adsorption and regeneration
processes in this work
was contained in an acrylic column (see Figure 3.5). Air entered
from the bottom of the
column at the inlet and moved upward through the zeolite,
exiting from the outlet at
the top.
40
-
To,8x12
Acrylic Tube
Sloped Insert
Silicone O-Ring
Inlet
200 x 200 Mesh Steel Cloth Disk
Outlet
Z=59 mmTo,60x65 @ 30 SLM
} Inlet Quick-Clamp Sub-Assembly
Z=19 mm
} Outlet Quick-Clamp Sub-Assembly
Z=54 mm
Z=49 mm
Z=39 mm
To,60x65 @ 25 SLMTo,60x65 @ 20 SLM
To,60x65 @ 15 SLM
Figure 3.5: Cross section of the adsorption column.
Samples of zeolite rested on the steel cloth disc which was
glued between the sloped
insert and the acrylic tube collar as part of the inlet
sub-assembly (see Figure A2).
41
-
Thermocouples were located axially within the column and it was
assumed that there
was no radial temperature gradient. The outlet temperature for
the fixed bed (T8x12)
was measured by a T-type thermocouple just above the height of a
25 g sample of 8x12
mesh zeolite at Z = 19 mm. Similarly, the outlet temperatures of
the fluidized bed
(T60x65) was measured by thermocouples at the peak bubble height
of a 25 g sample of
60x65 mesh zeolite fluidized at each flow rate (Zf ). A
different acrylic column was used
for each flow rate of the fluidized bed.
The top of the acrylic column was glued inside the outlet quick
clamp-collar sub
assembly, while the bottom portion was sealed by a silicone
o-ring and could be separated
from the inlet quick-clamp sub-assembly for cleaning and
maintenance of the column.
The inlet and outlet quick-clamp sub-assemblies interfaced with
the rest of the plumbing
of the apparatus by way of matching machined parts (see Appendix
A for engineering
drawings).
The shape of the fixed bed of 8x12 mesh zeolite was approximated
as a truncated
cone (Figure 3.6) and its volume was calculated using Equation
3.1, where Z is the
height of the static bed at 19 mm.
Figure 3.6: Dimensions of fixed bed for calculating volume
(units in mm).
Vfixed =1
3π(r21 + r1r2 + r
22
)Zo
=1
3π((23.88 mm)2 + (23.88 mm)(28.7 mm) + (28.7 mm)2
)(19 mm)
= 41.4 cm3
(3.1)
42
-
This volume was used to calculate the energy density of the
column for the fixed
bed. The shape of the fluid bed of 60x65 mesh zeolite was
considered to be that of the
fixed bed plus the volume of a cylinder the height of the
bubbles above the static height
of the fixed bed for a particular flow rate (Zf − Zo) per Figure
3.7 and Equation 3.2.
Figure 3.7: Dimensions of fluid bed for calculating volume
(units in mm).
Vfluid = Vfixed + πr22(Zf − Zo) (3.2)
Using Equation 3.2, the volume occupied by the fluidized
material in the column
at each flow rate was calculated, and the volumes indicated in
Table 3.1 were used to
calculate the energy density of the fluid bed.
Table 3.1: Volume of fluid bed in column at each flow rate.
Flow Rate (L/min) Zf (mm) Volume (cm3)
10 0 41.4
15 39 93.2
20 49 119
25 54 132
30 59 145
43
-
Figure 3.8 shows the height of the 60x65 mesh material fluidized
at 25 L/min in
comparison to the fixed bed
Figure 3.8: A 25 g sample of 8x12 zeolite in a fixed bed (left)
and 60x65zeolite fluidized at 25 L/min (right).
3.1.4 Instrumentation
The experimental setup shown in Figure 3.1 was equipped with the
following instru-
mentation to measure the adsorption and regeneration
processes:
• three pressure transducers with a compound range of -14.7 to
85 psig and an
accuracy of ±1% of the full-scale output (FSO)
• two mass flow controllers (MFC) with a range of 0 to 100
standard L/min (SLM),
and an accuracy of ±1.5% FSO (±1.5 SLM)
• two high-temperature relative humidity transmitters (RH) with
an accuracy of
±0.05% RH/◦C for -40 to 150◦C (±2% RH for 3 to 95% at 25◦C) for
relative
humidity and ±0.5◦C for temperatures of -40 to 180◦C
• T-type thermocouples within the adsorption column to measure
the outlet tem-
perature of the fixed and fluidized adsorbent beds with an
uncertainty of ±0.49◦C
44
-
A more detailed description of each instrument and the
uncertainty analysis applied
to each experiment is available as Appendix D.
3.1.5 Data Acquisition
To monitor all of the instrumentation installed, as well as to
provide the control
signals during experimental tests to the MFC and in-line air
heater, an existing National
Instruments NI cRIO 9024 CompactRIO chassis was wired to all of
the devices and
instruments through swappable I/O modules (cards) that make
specific measurements
or perform control functions. The configuration of the cards
installed in the NR cRIO
9024 are listed in Table 3.2.
Table 3.2: Configuration of I/O modules for the NI cRIO
9204.
I/OModule
FunctionNumber
of Modules
TotalNumber
of ChannelsPurpose
NI 9214Isothermal
Thermocouple Input1 16
measurethermocouples
NI 92074-20 mA and 0-10 V
Analog Input1 16
meausure RH,pressure, andflowrate
NI 92654-20 mA
Analog Output1 4 control flowrate
The data acquisition system communicated with a software called
LabVIEW that
allowed the signals to be read into and sent out from a program
referred to as a virtual
instrument (VI).
3.1.6 Inlet Humidity PID Control
Using the measured relative humidity at the inlet as feedback
for controlling the wet
and dry MFCs, it was possible to maintain a set relative
humidity at the inlet of the
adsorption column. Figure 3.9 shows the response of this control
as the inlet relative
humidity value of the LabVIEW VI was set to 30, 50, and 70% at a
total flow rate of
25 L/min. The flow rate of the wet and dry MFC are indicated on
the primary y-axis,
and the measured RH is indicated by the secondary y-axis.
45
-
Dry MFCWet MFCInlet Relative Humidity
Flo
w R
ate
(L/m
in)
5
7.5
10
12.5
15
17.5
20
22.5
25
27.5
30R
elative Hum
idity (%)
0
10
20
30
40
50
60
70
80
90
100
Time (min)0 10 20 30 40 50
Figure 3.9: Response of MFCs to change in inlet relative
humidity signal.
A plunge in the measured RH of the inlet air was a typical
response to the the start
of the VI and is why the column is bypassed for the first 5 min
of every experiment.
As the flows of the wet and dry MFC converge over the first 20
min, the measured
relative humidity of the mixed flows increases and stabilizes at
each setpoint value from
30→ 50→ 70% RH. Likewise as the two flows begin to diverge at 30
mins, the measured
inlet relative humidity decreases from 70 back to 30% RH.
3.2 Material Preparation
The TCM chosen for this work was a zeolite 13X molecular sieve
adsorbent purchased
from Delta Adsorbents (see Appendix B for manufacturer
specifications and MSDS). The
material was received as a 5 lb pail of 8x12 mesh beads (1.41 -
2.38 mm). To obtain
a 60x65 (Tyler) mesh adsorbent sample for bubbling fluidization,
a mortar and pestle
were used to crush portions of the 8x12 mesh material, which
were then shaken in a
stack of No. 50, 60, and 70 U.S. size sieves for 30 minutes. The
particles collected from
46
-
the No. 70 (65 Tyler mesh) sieve were assumed to be smaller than
the No. 60 (60 Tyler
mesh) sieve size and therefore have a particle diameter of 212 -
250 µm.
Per the manufacturer’s specifications, the zeolite was packaged
with ≤1.5 % wt.
moisture, which could be up to 1.5 g of water for a 100 g
sample. Prior to weighing
a sample for adsorption, batches of each mesh size were
dehydrated in a kiln at 350◦C
for 4 hours. Samples of 25 g were measured out while the
material was still hot from
coming out of the kiln, and placed in a sealed container labeled
according to the test it
was to be used for. Thes